A bit of history
The Machine Learning for Smart Mobility (MLSM) group was created in 2016, building on earlier initiatives of the same team in MIT/ITSLab, SMART (Singapore-MIT Alliance for Research and Technology), and University of Coimbra. These about 10 prior years of intensive research brought us expertise in Machine Learning, Transport Engineering, Behavior Modeling, Simulation, Optimization, and Data Collection.
It was not by chance that MLSM found its home in the Transport Division of DTU. Its research focus lies in the combination of the other two Transport Division groups (Demand Modeling and Network Modeling), particularly by applying Machine Learning techniques to Mobility research.
Our mission is to develop cutting edge research in both Machine Learning and Transport Research fields. Our belief is that, by its complexity and real-world impact, mobility is an excellent area to develop new ML methodologies, eventually applicable to other fields. We also believe that, only when one is truly engaged with the domain (and its theory), it is possible to advance science and technology in a consistent way. It is for this reason that our publication record combines articles in top Machine Learning venues (such as IEEE Transactions on Pattern Analysis and Machine Intelligence, Pattern Recognition, AAAI, ICML, and many more to come) as well as in the main Transport Research ones (IEEE Transactions on ITS, Transport Research Part C, Journal of ITS and others).
The two other fundamental MLSM components to mention relate with Simulation and Optimization. The former plays a key role in transport and urban models, because it is the natural way to explicitly include behavior models (demand) and try new smart mobility services (supply). Finally, optimization is an essential ingredient for the adaptive transport system that we need in our world. While we also have expertise in the field, we strongly rely on collaboration, towards the ultimate goal of predictive optimization.
Besides research, the group is committed to cutting edge education, and is responsible for several courses at DTU, covering topics such as Model-based Machine Learning, Data Sciences, Intelligent Transportation Systems and Smart Cities. Learn more about our courses here.